Improving IdSay: A Characterization of Strengths and Weaknesses in Question Answering Systems for Portuguese

被引:0
|
作者
Carvalho, Gracinda [1 ,2 ,3 ]
de Matos, David Martins [2 ,4 ]
Rocio, Vitor [1 ,3 ]
机构
[1] Univ Aberta, Rua Escola Politecn 147, P-1269001 Lisbon, Portugal
[2] INESC ID, L2f, R Alves Redol 9, P-1000 Lisbon, Portugal
[3] Univ Nova Lisboa, Fac Ciencias & Tecnol, Ctr Informat Technol, P-1200 Lisbon, Portugal
[4] Univ Tecn Lisboa, Inst Super Tecn, Lisbon, Portugal
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
IdSay is a Question Answering system for Portuguese that participated at QA@CLEF 2008 with a baseline version (IdSayBL). Despite the encouraging results, there was still much room for improvement. The participation of six systems in the Portuguese task, with very good results either individually or in an hypothetical combination run, provided a valuable source of information. We made an analysis of all the answers submitted by all systems to identify their strengths and weaknesses. We used the conclusions of that analysis to guide our improvements, keeping in mind the two key characteristics we want for the system: efficiency in terms of response time and robustness to treat different types of data. As a result, an improved version of IdSay was developed, including as the most important enhancement the introduction of semantic information. We obtained significantly better results, from an accuracy in the first answer of 32.5% in IdSayBL to 50.5% in IdSay, without degradation of response time.
引用
收藏
页码:1 / +
页数:2
相关论文
共 50 条
  • [1] IdSay: Question Answering for Portuguese
    Carvalho, Gracinda
    de Matos, David Martins
    Rocio, Vitor
    EVALUATING SYSTEMS FOR MULTILINGUAL AND MULTIMODAL INFORMATION ACCESS, 2009, 5706 : 345 - +
  • [2] Improving the Robustness of Question Answering Systems to Question Paraphrasing
    Gan, Wee Chung
    Ng, Hwee Tou
    57TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2019), 2019, : 6065 - 6075
  • [3] Improving Question-Answering for Portuguese Using Triples Extracted from Corpora
    Rodrigues, Ricardo
    Gomes, Paulo
    COMPUTATIONAL PROCESSING OF THE PORTUGUESE LANGUAGE (PROPOR 2016), 2016, 9727 : 25 - 37
  • [4] Pseudo-labeling with transformers for improving Question Answering systems
    Kuligowska, Karolina
    Kowalczuk, Bartlomiej
    KNOWLEDGE-BASED AND INTELLIGENT INFORMATION & ENGINEERING SYSTEMS (KSE 2021), 2021, 192 : 1162 - 1169
  • [5] RAPPORT - A Portuguese Question-Answering System
    Rodrigues, Ricardo
    Gomes, Paulo
    PROGRESS IN ARTIFICIAL INTELLIGENCE-BK, 2015, 9273 : 771 - 782
  • [6] Question Answering for Portuguese: How Much Is Needed?
    Wilkens, Rodrigo
    Villavicencio, Aline
    ADVANCES IN ARTIFICIAL INTELLIGENCE - SBIA 2010, 2010, 6404 : 173 - 182
  • [7] Priberam's question answering system for Portuguese
    Amaral, Carlos
    Figueira, Helena
    Martins, Andre
    Mendes, Afonso
    Mendes, Pedro
    Pinto, Claudia
    ACCESSING MULTILINGUAL INFORMATION REPOSITORIES, 2006, 4022 : 410 - 419
  • [8] A Portuguese Dataset for Evaluation of Semantic Question Answering
    de Araujo, Denis Andrei
    Rigo, Sandro Jose
    Quaresma, Paulo
    Muniz, Joao Henrique
    COMPUTATIONAL PROCESSING OF THE PORTUGUESE LANGUAGE, PROPOR 2020, 2020, 12037 : 217 - 227
  • [9] Evaluating the strengths and weaknesses of large language models in answering neurophysiology questions
    Shojaee-Mend, Hassan
    Mohebbati, Reza
    Amiri, Mostafa
    Atarodi, Alireza
    SCIENTIFIC REPORTS, 2024, 14 (01):
  • [10] Question/Answering Systems
    Visser, Ubbo
    KUNSTLICHE INTELLIGENZ, 2012, 26 (02): : 191 - 195